18 hours ago, cosmos8520 said:

I was able to resolve for standard 02 A1111 and 02.forge (the only two I tried) by running the following three commands in docker console AFTER the error appeared in the log.  All were mentioned in above posts but took me a few tries before I got it right.

Executed in order:

wget https://bootstrap.pypa.io/get-pip.py;python3.10 get-pip.py

apt update;apt install python3-packaging

pip install setuptools==69.5.1

 

thank you i looked i swear

5 hours ago, rguinn said:

thank you i looked i swear

Unfortunately, it's band aid.  I have to do this on every launch.  Hopefully the Docker maintainer (or whoever is responsible for the error) issues a fix soon.

Seems to be fixed now on the sd and sd forge repos. Just relaunch the docker and it'll pull the most recent repo

How is everyone managing the size of their container? Mine is currently at 28gb. Constantly increasing the size of docker.img isn't the correct way I know.

 

17 hours ago, sem1845 said:

How is everyone managing the size of their container? Mine is currently at 28gb. Constantly increasing the size of docker.img isn't the correct way I know.

 

This happened to me as well but the issue was that I was deleting stuff which was being moved to the container Trash folder. You can check if it's the same case for you by emptying Trash:
 

find ~/.local/share/Trash/* -delete

Another thing you can do is make sure the path allocations are correct and that you're not installing models etc to the docker.img but using appdata. 

Sorry didn't had much time recently.
On the latest version the bug preventing A1111 and to launch should be fixed.
I also fixed the github workflow that didn't update the "latest" version.
 

On 8/3/2024 at 2:07 AM, xikobe said:

This happened to me as well but the issue was that I was deleting stuff which was being moved to the container Trash folder. You can check if it's the same case for you by emptying Trash:
 

find ~/.local/share/Trash/* -delete

Another thing you can do is make sure the path allocations are correct and that you're not installing models etc to the docker.img but using appdata. 

I haven't deleted anything and can't find that location in the docker container.

 

This is my only mount for the container

/config-------/mnt/user/appdata/stable-diffusion

I you want to try Flux on the last forge you may encounter this error :
TypeError: 'NoneType' object is not iterable

the solution is to uninstall xformers. The easiest way to do so is to :
-remove xformers from the parameters.forge.txt file
-clean venv and reinstall dependencies but deleting this file : "Delete_this_file_to_clean_virtual_env_and_dependencies_at_next_launch"

11 hours ago, Holaf said:

I you want to try Flux on the last forge you may encounter this error :
TypeError: 'NoneType' object is not iterable

the solution is to uninstall xformers. The easiest way to do so is to :
-remove xformers from the parameters.forge.txt file
-clean venv and reinstall dependencies but deleting this file : "Delete_this_file_to_clean_virtual_env_and_dependencies_at_next_launch"

If you're unfamiliar with the location of the parameters file, it's in the folder you set as your UI path. Then you go into the Automatic1111 folder and you'll see the parameters txt file there.

Can I just put in a request for this flavour of image generation?

 

https://github.com/rupeshs/fastsdcpu/

 

Thanks for readying.

 

Oddsodz

Hi @superboki 

 

Thanks for this template.

 

After successfully installing and running the Automatic1111 (standalone) docker template, i wanted to give yours a go (as it offers more  flexibility and choice of all UIs).

I'm getting the following error on loop (from logs) : 

 

App is starting!
Reading package lists...
Building dependency tree...
Reading state information...
Package 'python3' is not installed, so not removed
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
Reading package lists...
Building dependency tree...
Reading state information...
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
/entry.sh: line 18: /03: No such file or directory
/entry.sh: line 19: /03: No such file or directory
/entry.sh: line 20: /config/scripts/03: No such file or directory
/entry.sh: line 21: /config/scripts/03: No such file or directory
error when launching WebUI

 

 

Where do i need to look to fix this ? (i made a mistake at first, i defined the UI model as "03", but then edited the template and made it "03 - InvokeAI" (maybe this is the error?)

 

Thanks

 

 

hello @Yagami I'll try to do a clean install to see if I have the same kind of error.

 

@OddsodzAllMighty I'll take a look at it.

BTW, for those interested I add Fluxgym to the container as option 71

@Yagami everything looks fine on my side.
and effectiveley you just have to set the variable to 03 to select InvokeAI :

 

Can't seem to get this to work. 

 

When I give a prompt it fails with this error:

Error: Could not load the stable-diffusion model! Reason: cannot import name 'cached_download' from 'huggingface_hub' (/config/01-easy-diffusion/installer_files/env/lib/python3.8/site-packages/huggingface_hub/__init__.py)

@Holaf

 

facefusion does not work anymore. they changed their entry point python file name and also parameters it seems

On 10/31/2024 at 6:15 AM, omnomasaur said:

Can't seem to get this to work. 

 

When I give a prompt it fails with this error:

Error: Could not load the stable-diffusion model! Reason: cannot import name 'cached_download' from 'huggingface_hub' (/config/01-easy-diffusion/installer_files/env/lib/python3.8/site-packages/huggingface_hub/__init__.py)

Easy diffusion isn't under active devellopment. You can either wait for them to fix this bug or try another WebUI.

I you really need it, I found a dirty fix here : https://github.com/easydiffusion/easydiffusion/issues/1851
you can edit this file inside the container : /easy-diffusion/installer_files/env/lib/python3.8/site-packages/diffusers/utils/dynamic_modules_utils.py and remove cached_download from the import line.

Hello, im getting the error when running Flux in InvokeAI. Anyone know how to fix this? Potentially related to a transformers bug, seems to be related to torch<2.4. According to this random github: https://github.com/QwenLM/Qwen2-VL/issues/12

 

"AttributeError: module 'torch.nn.functional' has no attribute 'rms_norm'"

 

Here's the logs:

[2024-12-16 17:59:09,518]::[InvokeAI]::ERROR --> Error while invoking session 2799580b-6fcb-4eef-afaa-cced045c1966, invocation 0d6897e7-aa55-4846-9c04-7fa4d01387c3 (flux_denoise): module 'torch.nn.functional' has no attribute 'rms_norm'
[2024-12-16 17:59:09,518]::[InvokeAI]::ERROR --> Traceback (most recent call last):
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/app/services/session_processor/session_processor_default.py", line 129, in run_node
    output = invocation.invoke_internal(context=context, services=self._services)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/app/invocations/baseinvocation.py", line 300, in invoke_internal
    output = self.invoke(context)
             ^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/app/invocations/flux_denoise.py", line 138, in invoke
    latents = self._run_diffusion(context)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/app/invocations/flux_denoise.py", line 334, in _run_diffusion
    x = denoise(
        ^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/backend/flux/denoise.py", line 73, in denoise
    pred = model(
           ^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1520, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/backend/flux/model.py", line 125, in forward
    img, txt = CustomDoubleStreamBlockProcessor.custom_double_block_forward(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/backend/flux/custom_block_processor.py", line 78, in custom_double_block_forward
    img, txt, img_q = CustomDoubleStreamBlockProcessor._double_stream_block_forward(
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/backend/flux/custom_block_processor.py", line 35, in _double_stream_block_forward
    img_q, img_k = block.img_attn.norm(img_q, img_k, img_v)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1520, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/backend/flux/modules/layers.py", line 79, in forward
    q = self.query_norm(q)
        ^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1520, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/invokeai/backend/flux/modules/layers.py", line 69, in forward
    return torch.nn.functional.rms_norm(x, self.scale.shape, self.scale, eps=1e-6)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: module 'torch.nn.functional' has no attribute 'rms_norm'

 

Is it possible to add this to Open WebUI? If not is that something that might be available in the future?

 

Thanks!

@Bunger try to delete the file named "Delete_this_file_to_clean_virtual_env_and_dependencies_at_next_launch"
It will reinstall all dependencies and you should have torch 2.4.1 after that.

@ZiRiuS Sorry I don't understand want you want to do with Open WebUI

Thank you @Holaf, it looks like deleting "Delete_this_file_to_clean_virtual_env_and_dependencies_at_next_launch" puts it in to a loop. I'm not sure if its iterating through options or a true loop. This keeps popping up and it seems to be attempting to download something that doesn't and is running out of space? Seems like I may have to allocate more disk? 
 

ERROR: Exception:
Traceback (most recent call last):
  File "/config/03-invokeai/env/lib/python3.11/site-packages/pip/_vendor/urllib3/response.py", line 438, in _error_catcher
    yield
  File "/config/03-invokeai/env/lib/python3.11/site-packages/pip/_vendor/urllib3/response.py", line 561, in read
    data = self._fp_read(amt) if not fp_closed else b""
           ^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/pip/_vendor/urllib3/response.py", line 527, in _fp_read
    return self._fp.read(amt) if amt is not None else self._fp.read()
           ^^^^^^^^^^^^^^^^^^
  File "/config/03-invokeai/env/lib/python3.11/site-packages/pip/_vendor/cachecontrol/filewrapper.py", line 102, in read
    self.__buf.write(data)
  File "/config/03-invokeai/env/lib/python3.11/tempfile.py", line 500, in func_wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 28] No space left on device

During handling of the above exception, another exception occurred:

 

logs.txt

Yea got it fixed, needed to clean up some other containers. Thanks for the help

On 12/29/2023 at 12:14 PM, domrockt said:

Can the Intel ARC770 be supported? Afaik there are Modifications to allow that. Your 

 

regards Dom

 

(i got it working with an regular Docker command and Portainer) 

(converted a docker to Unraid)

 

@domrockt Do you mind sharing your Portainer stack (docker compose)? I am trying to run Stable Diffusion in Portainer and finding a lot of difficulties locating a good image that works. Thanks in advance.

On 12/30/2024 at 7:27 PM, ebuy05 said:

 

@domrockt Do you mind sharing your Portainer stack (docker compose)? I am trying to run Stable Diffusion in Portainer and finding a lot of difficulties locating a good image that works. Thanks in advance.

 

i dont have the arc 770 anymore it was horrible to use with AI  and bought 2* 3090 as replacement.

i searched for „Arc770 stable Diffusion docker“ and Spun up that Image with Portainer.

 

*edit* found it nuullll/ipex-arc-sd:latest *edit*

 

i had alot of issues with Plugins and Sorts inside the docker. 

 

Edited January 12, 20251 yr by domrockt

Hi all. 

How are you all setting up things like Kohya with this container? I mean as far as storage locations? My cache drive is only 1tb so having AI models and or even the outputs on it is not going to work.. The last time I tried to configure an array drive instead, I kept getting errors about system incorrect configurations...  Any thoughts? 

Does anyone know which file I need to edit to make the 70-kohya install pull a branch other than main? Main is way behind, I need to train flux and it has yet to be merged into the main branch. So I need to switch to sd3-flux.1 branch. But where is this file and how to change it to pull that?